基于机器视觉的果园导航中线提取算法研究

为实现果园作业自动化,对基于机器视觉的果园自然情况下的导航路径拟合方法进行了研究。针对果园环境特点,基于HSV色彩模型和最大类间方差法对果园果树树行特征进行了增强。同时,使用Hough变换对果树树行线进行了拟合,并最终获得了相邻两树行的中线作为导航路径。最终试验结果表明:该方法能有效克服行间生草、光照阴影和修剪枯枝的干扰,可提取果园果树树行线并拟合出果园导航中线。

Navigation Line Extraction Based on Machine Vision in Orchard

In order to realize automation in orchard .This study provides a kind of method about navigation path fitting method in orchard environment for the machine vision navigation .According to the orchard environment ,HSV color model and Ostu were used to enhance the features of fruit tree line in orchard .Hough transform was Used to detect fruit tree row lines, and the midline , as a navigation path , was acquired between the two adjacent tree rows .Finally the experimental results show that this method can effectively extract the orchard fruit tree row lines and obtain the navigation line with overcoming the interference of grass , shadow and deadwood .